Training in progress, step 140

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2f12152976e2acde537479871935ba50770985026b0340f03ee13cdb679b503
+oid sha256:cb985399e15c4869e7f3bdeb8c3424817fa18edc02e72cbd715cbdac0a009bef
 size 6627156248

logging.jsonl

@@ -141,3 +141,14 @@
 {"loss": 0.36123657, "grad_norm": 0.19869718, "learning_rate": 4.711e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003305, "rewards/chosen": -0.03344727, "rewards/rejected": -0.08837891, "rewards/accuracies": 0.5, "rewards/margins": 0.05499268, "logps/rejected": -0.88378906, "logps/chosen": -0.33422852, "logits/rejected": -0.38818359, "logits/chosen": -0.60449219, "nll_loss": 0.33422852, "log_odds_ratio": -0.68121338, "log_odds_chosen": 0.76708984, "epoch": 0.45744681, "global_step/max_steps": "129/846", "percentage": "15.25%", "elapsed_time": "10h 50m 17s", "remaining_time": "2d 12h 14m 22s"}
 {"loss": 0.37228394, "grad_norm": 0.11339411, "learning_rate": 4.704e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003311, "rewards/chosen": -0.03164673, "rewards/rejected": -0.03961182, "rewards/accuracies": 0.75, "rewards/margins": 0.00796509, "logps/rejected": -0.39624023, "logps/chosen": -0.31665039, "logits/rejected": -0.38671875, "logits/chosen": -0.43017578, "nll_loss": 0.31665039, "log_odds_ratio": -0.60546875, "log_odds_chosen": 0.26025391, "epoch": 0.46099291, "global_step/max_steps": "130/846", "percentage": "15.37%", "elapsed_time": "10h 54m 11s", "remaining_time": "2d 12h 3m 7s"}
 {"eval_loss": 0.38319752, "eval_runtime": 568.3988, "eval_samples_per_second": 0.491, "eval_steps_per_second": 0.062, "eval_rewards/chosen": -0.03471157, "eval_rewards/rejected": -0.34204799, "eval_rewards/accuracies": 0.77142859, "eval_rewards/margins": 0.30736259, "eval_logps/rejected": -3.4203124, "eval_logps/chosen": -0.34718192, "eval_logits/rejected": -0.40456194, "eval_logits/chosen": -0.52040666, "eval_nll_loss": 0.34737724, "eval_log_odds_ratio": -0.39023611, "eval_log_odds_chosen": 3.7414341, "epoch": 0.46099291, "global_step/max_steps": "130/846", "percentage": "15.37%", "elapsed_time": "11h 3m 40s", "remaining_time": "2d 12h 55m 18s"}
+{"loss": 0.3994751, "grad_norm": 0.21607526, "learning_rate": 4.698e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003263, "rewards/chosen": -0.03521729, "rewards/rejected": -0.62878418, "rewards/accuracies": 1.0, "rewards/margins": 0.59417725, "logps/rejected": -6.29785156, "logps/chosen": -0.35205078, "logits/rejected": -0.53710938, "logits/chosen": -0.4921875, "nll_loss": 0.35107422, "log_odds_ratio": -0.13183594, "log_odds_chosen": 6.92041016, "epoch": 0.46453901, "global_step/max_steps": "131/846", "percentage": "15.48%", "elapsed_time": "11h 8m 45s", "remaining_time": "2d 12h 50m 5s"}
+{"loss": 0.42602539, "grad_norm": 0.46306911, "learning_rate": 4.691e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003268, "rewards/chosen": -0.03283691, "rewards/rejected": -0.86950684, "rewards/accuracies": 1.0, "rewards/margins": 0.83569336, "logps/rejected": -8.69921875, "logps/chosen": -0.32861328, "logits/rejected": -0.50488281, "logits/chosen": -0.58447266, "nll_loss": 0.32861328, "log_odds_ratio": -0.12402344, "log_odds_chosen": 9.47021484, "epoch": 0.46808511, "global_step/max_steps": "132/846", "percentage": "15.60%", "elapsed_time": "11h 12m 50s", "remaining_time": "2d 12h 39m 25s"}
+{"loss": 0.38085938, "grad_norm": 0.19084617, "learning_rate": 4.685e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003276, "rewards/chosen": -0.0513916, "rewards/rejected": -0.8175354, "rewards/accuracies": 0.75, "rewards/margins": 0.7661438, "logps/rejected": -8.18359375, "logps/chosen": -0.51416016, "logits/rejected": -0.4375, "logits/chosen": -0.56396484, "nll_loss": 0.51513672, "log_odds_ratio": -0.40136719, "log_odds_chosen": 8.17089844, "epoch": 0.47163121, "global_step/max_steps": "133/846", "percentage": "15.72%", "elapsed_time": "11h 16m 18s", "remaining_time": "2d 12h 25m 36s"}
+{"loss": 0.35171509, "grad_norm": 0.08958192, "learning_rate": 4.678e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003291, "rewards/chosen": -0.04199219, "rewards/rejected": -0.11065674, "rewards/accuracies": 0.5, "rewards/margins": 0.06860352, "logps/rejected": -1.10644531, "logps/chosen": -0.41943359, "logits/rejected": -0.43115234, "logits/chosen": -0.59423828, "nll_loss": 0.41943359, "log_odds_ratio": -0.6411438, "log_odds_chosen": 0.95214844, "epoch": 0.4751773, "global_step/max_steps": "134/846", "percentage": "15.84%", "elapsed_time": "11h 18m 24s", "remaining_time": "2d 12h 4m 41s"}
+{"loss": 0.38513184, "grad_norm": 0.09886587, "learning_rate": 4.671e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003297, "rewards/chosen": -0.03173828, "rewards/rejected": -0.07641602, "rewards/accuracies": 0.75, "rewards/margins": 0.0446167, "logps/rejected": -0.76293945, "logps/chosen": -0.31762695, "logits/rejected": -0.31774902, "logits/chosen": -0.3684082, "nll_loss": 0.31689453, "log_odds_ratio": -0.63311768, "log_odds_chosen": 0.62011719, "epoch": 0.4787234, "global_step/max_steps": "135/846", "percentage": "15.96%", "elapsed_time": "11h 22m 5s", "remaining_time": "2d 11h 52m 22s"}
+{"loss": 0.34994507, "grad_norm": 0.12538408, "learning_rate": 4.665e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003301, "rewards/chosen": -0.04974365, "rewards/rejected": -0.13781738, "rewards/accuracies": 1.0, "rewards/margins": 0.0880127, "logps/rejected": -1.37890625, "logps/chosen": -0.49755859, "logits/rejected": -0.40209961, "logits/chosen": -0.37643433, "nll_loss": 0.49804688, "log_odds_ratio": -0.21289062, "log_odds_chosen": 1.56445312, "epoch": 0.4822695, "global_step/max_steps": "136/846", "percentage": "16.08%", "elapsed_time": "11h 26m 26s", "remaining_time": "2d 11h 43m 36s"}
+{"loss": 0.39938354, "grad_norm": 0.15636803, "learning_rate": 4.658e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003306, "rewards/chosen": -0.03649902, "rewards/rejected": -0.06018066, "rewards/accuracies": 0.5, "rewards/margins": 0.02368164, "logps/rejected": -0.6015625, "logps/chosen": -0.36474609, "logits/rejected": -0.44482422, "logits/chosen": -0.59765625, "nll_loss": 0.36425781, "log_odds_ratio": -0.54370117, "log_odds_chosen": 0.50878906, "epoch": 0.4858156, "global_step/max_steps": "137/846", "percentage": "16.19%", "elapsed_time": "11h 30m 26s", "remaining_time": "2d 11h 33m 11s"}
+{"loss": 0.43347168, "grad_norm": 0.18616471, "learning_rate": 4.652e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003309, "rewards/chosen": -0.03619385, "rewards/rejected": -0.1854248, "rewards/accuracies": 1.0, "rewards/margins": 0.14916992, "logps/rejected": -1.85449219, "logps/chosen": -0.36230469, "logits/rejected": -0.36791992, "logits/chosen": -0.34692383, "nll_loss": 0.36230469, "log_odds_ratio": -0.22433472, "log_odds_chosen": 2.20703125, "epoch": 0.4893617, "global_step/max_steps": "138/846", "percentage": "16.31%", "elapsed_time": "11h 34m 49s", "remaining_time": "2d 11h 24m 43s"}
+{"loss": 0.41183472, "grad_norm": 0.23319234, "learning_rate": 4.645e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003316, "rewards/chosen": -0.03393555, "rewards/rejected": -1.46806335, "rewards/accuracies": 0.75, "rewards/margins": 1.43373108, "logps/rejected": -14.67285156, "logps/chosen": -0.33935547, "logits/rejected": -0.36621094, "logits/chosen": -0.41186523, "nll_loss": 0.33911133, "log_odds_ratio": -0.29882812, "log_odds_chosen": 15.13476562, "epoch": 0.4929078, "global_step/max_steps": "139/846", "percentage": "16.43%", "elapsed_time": "11h 38m 24s", "remaining_time": "2d 11h 12m 19s"}
+{"loss": 0.34051514, "grad_norm": 0.1030272, "learning_rate": 4.639e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.00332, "rewards/chosen": -0.02624512, "rewards/rejected": -0.7204895, "rewards/accuracies": 0.5, "rewards/margins": 0.69430542, "logps/rejected": -7.20874023, "logps/chosen": -0.26245117, "logits/rejected": -0.47607422, "logits/chosen": -0.53564453, "nll_loss": 0.26318359, "log_odds_ratio": -0.47558594, "log_odds_chosen": 7.64453125, "epoch": 0.4964539, "global_step/max_steps": "140/846", "percentage": "16.55%", "elapsed_time": "11h 42m 35s", "remaining_time": "2d 11h 3m 2s"}
+{"eval_loss": 0.37720513, "eval_runtime": 569.7455, "eval_samples_per_second": 0.49, "eval_steps_per_second": 0.061, "eval_rewards/chosen": -0.03427734, "eval_rewards/rejected": -0.82667762, "eval_rewards/accuracies": 0.74285716, "eval_rewards/margins": 0.79263741, "eval_logps/rejected": -8.26568127, "eval_logps/chosen": -0.34282923, "eval_logits/rejected": -0.46145368, "eval_logits/chosen": -0.57289344, "eval_nll_loss": 0.34315011, "eval_log_odds_ratio": -0.35142821, "eval_log_odds_chosen": 8.68630028, "epoch": 0.4964539, "global_step/max_steps": "140/846", "percentage": "16.55%", "elapsed_time": "11h 52m 4s", "remaining_time": "2d 11h 50m 55s"}